can not train lora

[Zyj1moreT]

I download all files from huggingface，but I find this error ： ValueError: Cannot load local_pretrained_models/diffusion because down_blocks.0.attentions.0.transformer_blocks.0.attn2.to_k.weight expected shape tensor(..., device='meta', size=(320, 1280)), but got torch.Size([320, 768]). If you want to instead overwrite randomly initialized weights, please make sure to pass both low_cpu_mem_usage=False and ignore_mismatched_sizes=True. For more information, see also: https://github.com/huggingface/diffusers/issues/1619#issuecomment-1345604389 as an example.

It’s my config.json fault？ how can I solve it

[Zyj1moreT]

I find all my parameters passed to UNet2DConditionModel, but are not expected and will be ignored. I can run the process，but I can‘t train lora.

[Yujun-Shi]

Hello, it seems that your error message is not due to lora training, but because you cannot load the pretrained diffusion model. It's possible that you've downloaded the wrong pretrained models. Please make sure you downloaded Stable Diffusion 1.5 in this link: https://huggingface.co/runwayml/stable-diffusion-v1-5/tree/main

[Zyj1moreT]

I find the problem! thank U !